The Effect of Pleural Abrasion on the Treatment of Primary Spontaneous Pneumothorax: A Systematic Review of Randomized Controlled Trials

Pleural abrasion has been widely used to control the recurrence of primary spontaneous pneumothorax (PSP). However, controversy still exists regarding the advantages and disadvantages of pleural abrasion compared with other interventions in preventing the recurrence of PSP.The PubMed, Embase, and Cochrane Central Register of Controlled Trials databases were searched up to December 15, 2014 to identify randomized controlled trials (RCTs) that compared the effects of pleural abrasion with those of other interventions in the treatment of PSP. The study outcomes included the PSP recurrence rate and the occurrence rate of adverse effects.Mechanical pleural abrasion and apical pleurectomy after thoracoscopic stapled bullectomy exhibited similarly persistent postoperative air leak occurrence rates (p = 0.978) and 1-year PSP recurrence rates (p = 0.821), whereas pleural abrasion led to reduced residual chest pain and discomfort (p = 0.001) and a smaller rate of hemothorax (p = 0.036) than did apical pleurectomy. However, the addition of minocycline pleurodesis to pleural abrasion did not reduce the pneumothorax recurrence rate compared with apical pleurectomy (3.8% for both procedures) but was associated with fewer complications. There was no statistical difference in the pneumothorax recurrence rate between mechanical pleural abrasion and chemical pleurodesis with minocycline on either an intention-to-treat basis (4 of 42 versus 0 of 42, p = 0.12; Fisher exact test) or after exclusions (2 of 40 versus 0 of 42, p = 0.24; Fisher exact test). Pleural abrasion plus minocycline pleurodesis also did not reduce the pneumothorax recurrence rate compared with pleural abrasion alone (p = 0.055). Moreover, pleural abrasion plus minocycline pleurodesis was associated with more intense acute chest pain. The postoperative overall recurrence rate in patients who underwent staple line coverage with absorbable cellulose mesh and fibrin glue was similar to that with mechanical abrasion after thoracoscopic bullectomy (13.8% vs. 14.2%, respectively; p = 0.555), but staple line coverage resulted in less postoperative residual pain than mechanical abrasion (0.4% vs.3.2%; p